Older adults with type 2 diabetes have a higher risk of fractures, adding to the health burden of this disease. More frequent falls and perhaps reduced bone strength in those with diabetes are thought to be key contributing factors. However, it is not clear whether better control of diabetes is sufficient to reduce fracture risk or whether additional treatment for osteoporosis should be given. Observational studies suggest that improved glycemic control may reduce fracture risk by preventing falls, decreasing the severity of falls, and reducing bone loss. Previous trials have demonstrated that improved control reduces diabetic complications, particularly peripheral neuropathy and retinopathy, that are risk factors for falls. Intensive glycemic control may preserve bone strength through reduced levels of advanced glycation endproducts in the bone, improved bone cell function, and reduced bone loss. On the other hand, increased hypoglycemic episodes accompanying tight glycemic control could promote falls in the older diabetic population, resulting in an increased risk of fracture. In order to determine whether intensive glycemic control is useful as a prevention measure for fractures, falls and/or bone loss in older diabetic adults, we propose adding measures of fractures, falls and bone mineral density to the recently initiated clinical trial Action to Control Cardiovascular Risk in Diabetes (ACCORD). ACCORD will follow 10,000 middle-aged and older adults (average age 63 years) with type 2 diabetes for an average of 5.6 years to determine if cardiovascular events are reduced by intensive (A1C <6%) versus standard (A1C approximately 7.5%) glycemic control. We hypothesize that those randomized to intensive control will have a lower rate of fractures, less frequent falls, and reduced bone loss compared with those randomized to standard control. In 7,145 of the ACCORD participants, fractures and falls will be determined by self-report every 12 months, with reported fractures adjudicated centrally using medical records. Change in bone mineral density over two years at the hip and lumbar spine will be determined in a subsample (N=240) using dual energy absorptiometry (DEXA).